Update README.md

README.md

@@ -26,11 +26,18 @@ Hinglish-Top [Dataset](https://huggingface.co/datasets/WillHeld/hinglish_top) co
 - domain 
 
 ### Training
-|Epochs|Train Loss|
-|:------:|:----------:|
-|4th   |   0.251 |
-
-The model was trained only for 4 epochs due to the GPU limitations. The model will give far better results with 10 epochs
+|Epoch|Loss|
+|:--:|:--:|
+|1	|0.0485|
+|2	|0.00837|
+|3	|0.00812|
+|4	|0.0029|
+|5	|0.014|
+|6	|0.00748|
+|7	|0.0041|
+|8	|0.00543|
+|9	|0.00304|
+|10	|0.000574|
 
 ### Inference 
 ```python